Lydia Krifka-Dobes

I’m Lydia Krifka-Dobes—I work with Samir Timajchi as a computer-artist.

Computer: (99% open source software and hardware, 1% other)
Artist: (51% sound and 49% other mediums)

Studies and Background

Studies at Institute of Sonology
Gap Year at Conservator Evangelicky Academy
High School at Beethoven Gymnasium
Studied at School for Poetic Computation
Based in Den Haag and Berlin
Born in Austin, Texas

system_boot = {
   artist: "lydia_krifka-dobes",
   location: "berlin",
   status: "exploring_the_glitch",
   focus: ["sound", "identity", "technology"],
   collaborators: ["humans", "machines", "systems"]
}

About

I write about experimental music, art-technology hybrids, and the poetics of code. Collaborate with me! I’m always open to new projects and opportunities.

You can email me at opensourcereartwork@gmail.com if you want to discuss art, technology, identity, or anything in between. I’m happy to engage in mind-expanding conversations!

Current Research

Research Projects Overview

Our work operates at the intersection of thermal-acoustic interaction, **language preservation through computational methods, and **deep fake transformation using advanced machine learning pipelines.

1. Temperature & Sound Research

Overview

At the intersection of physiological response and artistic expression, this research explores how temperature changes affect our perception and experience of sound.

Sauna Project Development

Collaborating with Borealis Festival in Bergen, we’re pioneering new approaches to immersive sound experiences in extreme temperature environments.

Key Components:

Full-scale sauna installations
Cold water immersion environments
Temperature gradient zones for performance
Real-time acoustic monitoring systems

CHILLER Technology Integration

The NYU CHILLER sensor enables unprecedented tracking of audience physiological responses, opening new possibilities for interactive performance.

Technical Specifications:

20mm x 20mm polymer construction
High-precision skin texture monitoring
Real-time piloerection detection
Multiple sensor array capability

Goosebumps Research

Studying the relationship between temperature changes and sensory responses.
Combining ASMR sounds, brass registers, and rapid thermal shifts.
Exploring the potential of wind machines to induce shivers in localized zones (e.g., eyes and sinus areas).

Insights from Correspondence

Subjectivity of Goosebumps:

Responses to sonic or temperature stimuli vary widely between individuals.

Rapid Temperature Changes:

Goosebumps are more likely to occur with sudden shifts rather than at specific thresholds.

Relevant Works and Practices:

The Flicker by Tony Conrad
Philippe Rahm’s Convection Spaces
Native American sweat lodge practices, which inspired early SWANS performances
Sensory installations by Olafur Eliasson

2. Language Preservation Project

Overview

This project documents and analyzes the acoustic diversity of human languages through high-fidelity recordings, computational analysis, and creative reinterpretation.

Framework Structure

Our archive centers on The Oak and the Reed fable, chosen for its rich phonetic content and narrative structure.

Recording Architecture:

12 parallel sentences with identical semantic content
24-bit/96kHz audio capture
Controlled recording environments
Multiple takes per speaker
Ambient noise profiling
Room impulse response measurements

Language Collection

We are working with 21 languages across multiple language families, including endangered ones:

Austronesian:
Caucasian:
- Abkhazian
Language Isolates:
- Basque
Germanic:
- Bavarian
Creole Languages:
- Bislama
- Rodriguan Creole
Endangered Pacific Languages:
- Daakie
African Languages:
- Hausa
- Khoekhoe
- Yoruba
Asian Languages:
- Malayalam
- Vietnamese
Mayan Family:
- Yucatec Maya

Technical Implementation

Segmentation Pipeline:

Using WhisperAI for sentence-level segmentation.
Phonetic alignment with Web-MAUS.

Naming Conventions:
[LANG]-[SENT]-[WORD]-[SYLL]-[PHON]-[SAMPA].wav Example: ABKH-3-4-2-3-O.wav Phonetic Analysis Tools:

Vowel space mapping
Consonant cluster analysis
Prosodic feature extraction
Tonal pattern recognition

Main Composition

Inspired by Trevor Wishart’s Globalalia, the main composition reimagines the Aesop fable through:

Strong, rigid oak tree: low sonority (voiceless stops)
Weak, flexible reeds: high sonority (vowels)
Water and wind as dynamic, sonic elements

3. Deep Fake Transformation Project

Overview

This project reimagines deep fake technology as an artistic tool for interactive, real-time data morphing across mediums like audio and video.

Binary Search Trees & Granular Synthesis

Concept:

Binary Search Trees (BST): Organize data into sorted nodes based on comparisons.
Granular Synthesis: Breaks audio into tiny “grains” for micro-level manipulation.

Applications:

Sorting audio by frequency centroid or RMS value.
Organizing video data by brightness or edge sharpness.
Morphing data between mediums in real time.

Procedure

Audio Transformation:
- Sorting samples by pitch or amplitude.
- Creating evolving sonic landscapes.
Visual Transformation:
- Sorting pixels by color intensity or spatial frequency.
- Dynamic texture generation.
Interactive Control:
- Multi-touch interfaces for real-time manipulation.
- OSC and MIDI integration for external device control.

Technical Challenges

Processing Large Datasets:

Efficient real-time analysis of audio and video.
Integration of OpenCV for computer vision tasks.

Creative Opportunities:

Combining granular synthesis with dynamic particle systems.
Leveraging Perlin noise for generative meshes.

Ethical Considerations

Consent & Transparency:

Explicit consent for data usage in deep fake creation.
Ensuring viewers understand the artificial nature of the work.

Artistic Responsibility:

Using deep fake technology to provoke critical thought on identity, **authenticity, and **perception rather than deception.

Contact

For collaboration inquiries, reach out at opensourcereartwork@gmail.com.